What Are Conflict Management Styles?
Conflict management styles are recognizable patterns in how people respond when their needs, goals, values, or interpretations seem incompatible with those of another person. The five-style framework distinguishes avoiding, accommodating, competing, compromising, and collaborating. Its central question is how much a person attempts to satisfy their own concerns and how much they attempt to satisfy the other party's concerns in a particular dispute. These dimensions offer a practical vocabulary for describing responses to disagreement, comparing their consequences, and deciding when a different approach might be more effective.
A conflict style refers to a response tendency or strategy. It is not a psychiatric diagnosis, a fixed personality category, or a reliable moral ranking. The same person can avoid one disagreement, compete over another, and collaborate on a third. A response also depends on what the other participant does, who controls resources, whether disagreement is safe, and what would count as an acceptable outcome. The foundational five-mode instrument was developed to measure conflict-handling behavior rather than to diagnose personality or psychological health (Kilmann and Thomas, 1977).
The five styles answer different practical questions. Avoiding postpones or sidesteps engagement. Accommodating yields to another person's preferences. Competing attempts to secure one's own preferred outcome. Compromising seeks an acceptable exchange of concessions. Collaborating jointly investigates interests and alternatives in search of an outcome that substantially addresses both parties' concerns. Their usefulness depends on the issue, the relationship, time pressure, available alternatives, and freedom to participate.
This article examines the theoretical model, the evidence behind its measurement, the conditions under which each style can help or harm, and the limits of using a five-part taxonomy to understand real interpersonal conflict. For the broader psychology of how disagreements arise and develop, see the published Communication, Conflict and Social Skills research chapter.
The Two Dimensions Behind the Five-Style Model
Concern for one's own goals
The first dimension is often called assertiveness or concern for self. It indicates the degree to which a response pursues the participant's own interests, preferred outcomes, rights, or priorities. In the five-mode diagram, competing and collaborating represent relatively high concern for self, avoiding and accommodating relatively low concern for self, and compromising an intermediate position. This dimension describes the direction of behavior during a dispute. It should not be confused with assertiveness as a broader communication skill, self-esteem, extraversion, or willingness to dominate someone.
For example, a person who insists that a shared bill be calculated accurately may exhibit strong concern for self while communicating calmly and respectfully. Another person may express the same need in a hostile manner. The goal being pursued and the manner of expression are related but separate properties. Research has also treated conflict behavior in terms of activity and agreeableness rather than assuming that every interaction fits neatly on a single self-versus-other continuum (van de Vliert and Euwema, 1994).
Concern for the other person's goals
The second dimension is cooperativeness or concern for others: how actively one attempts to accommodate another party's interests. Collaborating and accommodating are relatively high on this dimension; competing and avoiding are relatively low; compromising is often placed in the middle. A cooperative behavior can express genuine concern, pragmatic interdependence, a wish to protect a relationship, or a calculated decision to make a concession. Observing a cooperative response does not establish the participant's motive.
The dimensions are conceptually independent. A person can pursue both sets of interests vigorously, yielding collaborating, or neither set actively, yielding avoiding. They can prioritize the other participant's interests through accommodating or their own through competing. Compromising approximates an intermediate strategy in which parties relinquish some demands to obtain an agreement. These definitions reflect the dual-concern tradition informing both the Thomas–Kilmann model and Rahim's related five-style framework (Rahim, 1983; Kilmann and Thomas, 1977).
The familiar two-axis diagram is an organizing model of conduct. It is not a direct measurement of two universal brain mechanisms or proof that five psychologically distinct types exhaust human responses. Responses can be indirect, mixed, temporally changing, or shaped by fear and unequal bargaining power. Dyadic research has identified additional patterns beyond the five familiar labels (Guerrero, 2020).
How the Model Developed
Early models of managerial and interpersonal conflict distinguished concern for outcomes from concern for relationships and emphasized that people could approach disagreement through several recognizable modes. Kenneth W. Thomas and Ralph H. Kilmann developed the Thomas–Kilmann Conflict Mode Instrument, commonly abbreviated TKI, during the 1970s. Their 1977 methodological paper described a forced-choice assessment of five modes designed partly to reduce the tendency for respondents to select whichever conflict behavior appears most socially desirable (Kilmann and Thomas, 1977).
M. Afzalur Rahim subsequently developed a measure of styles for handling interpersonal conflict in organizational settings. The corresponding labels are integrating, obliging, dominating, avoiding, and compromising. In everyday discussion, integrating is closely related to collaborating, obliging to accommodating, and dominating to competing. The labels overlap theoretically, but questionnaire scores obtained from different measures should not automatically be treated as interchangeable (Rahim, 1983; Thomas and Kilmann, 1978).
These models became especially influential in organizational education, negotiation training, and communication courses because they make alternative responses visible. The same explanatory economy also creates a limitation: five memorable categories can conceal the content of a dispute, the quality of a relationship, opportunities to change underlying constraints, and the consequences of each party's conduct. Contemporary studies still use five-style measures, while testing whether personality facets, interpersonal role, directness, and relational context explain variations that broad labels miss (Espinoza, O'Neill, and Donia, 2023; Guerrero, 2020).
Avoiding: Stepping Away, Delaying, or Declining Engagement
Avoiding describes low active pursuit of one's own immediate demands and low active engagement with the other participant's demands. It can take the form of leaving a heated exchange, postponing a conversation, changing the subject, not replying, or deciding that an issue does not require discussion. The same outward action can have sharply different meanings. Pausing to recover from intense emotion is different from repeatedly evading accountability; choosing not to engage with an intimidating person is different from refusing to address a mutual obligation.
A useful instance of avoiding occurs when the decision is minor, reliable information is missing, temporary delay will improve judgment, or the immediate interaction is too activated for productive discussion. Two people who discover that their conversation is becoming accusatory may agree to pause until a specific time and return to the issue. In this case, withdrawal is an explicit part of a larger plan, not a claim that the underlying disagreement has disappeared.
Avoidance becomes costly when urgent safety or fairness problems remain unattended, when one participant repeatedly cannot obtain a response, or when accumulated grievances transform a manageable problem into a persistent relational pattern. The risk is not inherent in silence alone; duration, explanation, mutual expectation, and the unresolved stakes matter. Observational research with couples shows that withdrawal is associated with relationship satisfaction, although the associations are generally smaller than the association between hostility and dissatisfaction, and observational associations do not establish a simple causal direction (Woodin, 2011).
One must also distinguish avoidance from protective non-engagement. In a threatening setting, declining a confrontation, preserving privacy, leaving a location, or obtaining independent support may be the most appropriate actions. The conventional advice to confront every issue directly presumes a freedom to disagree that some people do not have.
Accommodating: Yielding to Another Person's Preference
Accommodating means prioritizing another party's concern over one's own immediate preference. It includes voluntarily accepting a different choice, making a concession, allowing another person to lead, or treating their immediate need as more important than one's own. In Rahim's framework, the closely corresponding label is obliging. Accommodation can express generosity, reciprocal care, an informed judgment about relative stakes, or a practical choice when one person has substantially more at risk.
Imagine that two friends disagree about a meeting location. One person can travel easily while the other has mobility limitations and a difficult commute. The first friend may accommodate the second because the costs of the options are genuinely unequal. Calling this response low assertiveness without describing the circumstances obscures why yielding may be reasonable.
Accommodation can also be useful when a participant recognizes an error, lacks crucial information, or values preserving time and goodwill on an issue of modest importance. It need not imply inferiority, passivity, or an inability to state one's needs. A person can calmly explain that they understand the alternatives and have chosen to defer.
Its costs arise when concessions are expected rather than chosen, when the accommodating party cannot express a meaningful refusal, or when one-sided sacrifices repeatedly accumulate. An agreement obtained through fear or dependency is not equivalent to voluntary cooperation. In these circumstances, the analytic task is to investigate power, choice, and consequences rather than label someone an inherently accommodating personality.
Competing: Pursuing One's Own Outcome Firmly
Competing involves high pursuit of one's own concerns with comparatively little effort to satisfy the other participant's preferred outcome. It may appear as insisting on a position, claiming scarce resources, refusing a proposed trade-off, enforcing a rule, or using a legitimate decision-making authority. Dominating is a related term in Rahim's taxonomy. The label describes an orientation toward outcome; the conduct may be procedurally appropriate or coercive, respectful or hostile, depending on what occurs.
Competing can serve a legitimate function when an immediate decision is necessary, when legal or professional obligations are clear, when a participant must refuse an unsafe demand, or when a fundamental right or nonnegotiable boundary is at stake. A person who declines to falsify a report despite social pressure is asserting a position that should not be averaged with the alternative. The issue has a normative constraint beyond mutual preference.
Overreliance on competition can intensify defensiveness, reduce willingness to share information, damage trust, or produce short-term compliance rather than durable acceptance. Nevertheless, the reverse inference is also invalid: a person who speaks firmly or resists another's demand is not thereby abusive or psychologically unhealthy. Competing becomes ethically problematic when force, intimidation, threats, exploitation, or retaliatory control replace legitimate disagreement.
Laboratory and organizational findings about contention and negotiation involve defined incentives and settings. They cannot be generalized mechanically to family relationships or to coercive circumstances. A central empirical question is whether a tactic secures an outcome while preserving agency, safety, and an acceptable ongoing relationship.
Compromising: Trading Concessions for an Acceptable Agreement
Compromising attempts to reach an agreement in which participants adjust demands or surrender some preferred outcomes. The classic model locates it near the middle of concern for self and concern for others. Participants might divide a limited resource, alternate responsibilities, shorten a disputed schedule, or combine parts of competing proposals. A compromise can be reached rapidly when parties agree that each is making an acceptable concession.
Suppose housemates disagree about noise after 10 p.m. One wants silence beginning at 9 p.m.; the other wants to host visitors until midnight. A compromise might establish quiet hours at 10:30 on weekdays and a different arrangement on weekends. Whether the result works depends on sleep needs, responsibilities, the reliability of enforcement, and whether those affected can meaningfully consent.
The familiar split-the-difference solution is only one form of compromise. Parties can exchange concessions across different issues, sequence commitments, or agree on a reversible trial. These options often provide more flexibility than choosing the arithmetic midpoint. If the disputed issue involves a legal protection, basic dignity, safety, or another right, however, the midpoint between compliance and violation is not a fair solution merely because it is symmetric.
Compromise is attractive when time is limited, stakes are moderate, and discovering a fully integrated solution is impractical. Its limitation is that bargaining over stated positions can leave underlying needs unidentified. A short-term settlement is not proof that the distributive outcome is equitable or that the problem will remain solved.
Collaborating: Jointly Solving the Underlying Problem
Collaborating combines active pursuit of one's own concerns with sustained attention to the concerns of the other person. Participants clarify what matters, distinguish positions from underlying interests, share relevant information, generate alternatives, assess consequences, and establish a workable agreement. Rahim's integrating style is conceptually similar. Collaboration is an interactive process requiring genuine participation; one person's invitation to collaborate cannot guarantee a collaborative result.
Consider two colleagues who argue over ownership of a project. One needs authorship recognition; the other needs timely delivery and decision-making clarity. Instead of dividing the project title in half or having one colleague surrender entirely, they may specify complementary roles, clear attribution, milestones, and review criteria. This is an integrative proposal if it meaningfully addresses both sets of interests and rests on accurate information.
The strongest reason to explore collaboration is the possibility that a conflict framed as a single scarce choice contains multiple interests that can be satisfied in different ways. In a meta-analysis of negotiation studies, prosocial motives were associated with greater problem solving and higher joint outcomes under particular conditions involving resistance to yielding; those moderators matter, and the findings do not establish a universal advantage in every interpersonal setting (De Dreu, Weingart, and Kwon, 2000).
Collaboration requires attention, time, a tolerable emotional climate, adequate information, and some prospect that participants can make and honor commitments. High-stakes unequal power relationships, threats, concealed coercion, and bad-faith bargaining can defeat these conditions. Requests for unlimited openness may even expose a vulnerable participant to manipulation. For that reason, a collaborative invitation must be conditional on safety and free choice, not treated as an obligation to remain in a harmful exchange.
Comparing the Five Styles in the Same Disagreement
Take one repeated conflict: two adults share responsibility for caring for an older relative, but one repeatedly handles more appointments. Avoiding means postponing the conversation or disengaging from the disagreement. Accommodating means one person agrees to continue the uneven arrangement. Competing means a person firmly demands a particular distribution. Compromising means the parties trade appointments or divide them by a mutually acceptable formula. Collaborating means they investigate each person's availability, financial means, transport limitations, the relative's preferences, and possible outside assistance before designing an arrangement.
These five descriptions concern the conflict-handling process. None guarantees a just outcome. An avoidant pause may enable later problem solving; a competitive demand may insist on necessary fairness; a compromise may hide unequal constraints; and a superficially collaborative meeting may still privilege the person with more money or authority. The final evaluation must examine consequences and voluntariness alongside the label.
The styles also coexist over time. An exchange may begin with avoidance, shift to competition when a boundary is challenged, then move toward compromise or collaboration after missing information emerges. A style profile based on one conversation would therefore be an incomplete account of how either participant manages conflict in general.
Conflict Style, Conflict Outcome, and Relationship Quality
A conflict response can be evaluated in at least four ways: whether it achieves the immediate objective; whether the allocation or decision is substantively fair; whether it protects the relationship where continued interaction is desirable; and whether the process preserves each participant's autonomy and safety. These criteria need not move in the same direction. Winning a concession might accomplish a goal while weakening future trust. Agreeing quickly might preserve surface calm while leaving an obligation unaddressed. A lengthy collaborative process might improve understanding but fail to produce resources that do not exist.
Research on observed couple behavior provides evidence about the importance of actual interaction patterns, but it does not directly validate a five-style score as the cause of relationship satisfaction. In a meta-analysis of 64 studies involving 5,071 couples, Erica Woodin reported that hostility was robustly associated with lower satisfaction, while constructive problem-solving and intimacy behaviors were positively associated with satisfaction; distress and withdrawal showed smaller associations. These are correlational findings from studies that differed in setting, sample, and observation protocol (Woodin, 2011).
Related dyadic work complicates the idea that five modes exhaust the behavior people recognize in conflict. Guerrero examined reports from 256 romantic couples and identified a differentiated pattern including indirect fighting, while distinguishing neutral avoidance from more antagonistic forms of conflict behavior. The study supports caution about assuming that all indirect behavior is the same or that five labels necessarily have equivalent meaning across close relationships (Guerrero, 2020).
Team evidence adds another warning about simplistic prescriptions. A meta-analysis of task and relationship conflict found that both were negatively associated, on average, with team performance and member satisfaction, despite theories predicting that task disagreements might consistently help teams. Associations varied with task complexity and the extent to which task conflict became entangled with interpersonal conflict (De Dreu and Weingart, 2003). These findings address team conflict and should not be recast as proof that one individual's five-style score determines performance.
What Scientific Assessments Actually Measure
The Thomas–Kilmann Conflict Mode Instrument
The Thomas–Kilmann instrument uses forced-choice pairs of statements intended to describe alternative ways of responding to conflict. A respondent chooses which statement is closer to how they would behave. The original development paper explicitly investigated social-desirability bias, a problem that emerges when a respondent reports conduct regarded as admirable rather than conduct that accurately reflects their behavior. The forced-choice design reduced some aggregate social-desirability effects relative to comparison instruments, but did not eliminate individual response bias (Kilmann and Thomas, 1977).
The five scores are interdependent because selections allocate choices among modes. A higher relative score in one mode necessarily changes what can be observed in others; the scores are not five free-standing measurements of separate capacities. A high avoiding score means that avoiding was selected comparatively often within that test's response structure. It is not equivalent to an independently observed rate of avoidance in all everyday conflicts.
A comparison study by Thomas and Kilmann examined test–retest reliability, internal consistency, and convergent validity in a sample of 86 graduate management students. The authors characterized reliability as low to moderate and found variation in convergence across the instruments and modes (Thomas and Kilmann, 1978). The sample, design, and results limit sweeping claims that the instrument delivers a precise lifelong classification.
Rahim's framework and its distinct measurement logic
Rahim's original measure was designed to assess modes of handling interpersonal conflict, especially within role relationships. Its familiar labels overlap with the five-mode taxonomy but emphasize integrating, obliging, dominating, avoiding, and compromising. Depending on the version and context, a question can refer to behavior with a supervisor, subordinate, or peer. These distinctions matter because the other person's role changes perceived costs, obligations, authority, and the feasibility of disagreement (Rahim, 1983).
Evidence from organizational samples suggests that conflict handling varies with counterpart role. In a study of managers, respondents reported different patterns with superiors, subordinates, and peers. The study is informative about context sensitivity but should not be taken as a timeless cultural rule or causal effect of organizational rank (Rahim, 1986).
Self-report, observation, and ecological validity
Self-report questionnaires measure what participants say they generally do or prefer. Observational studies code what people display in a specific interaction. Partner reports measure how behavior is perceived by the other participant. Experimental negotiations manipulate incentives and information under controlled conditions, while diaries and repeated measurements can capture shifts across episodes. Each method answers a different question, and none automatically stands in for the others.
A person may endorse collaboration because it represents their values yet compete during intense time pressure. A partner may perceive a polite withdrawal as indifference, whereas the withdrawing person believes they are avoiding escalation. Neither report is meaningless; the discrepancy is information about the interaction. Stronger conclusions require triangulation, clear definitions of the unit of analysis, and attention to behavior over time.
A conflict-style assessment can therefore be used as a structured prompt for reflection, education, or a research hypothesis. It does not diagnose a disorder, determine whether someone is an ethical person, predict behavior with certainty, or establish whether a relationship is safe.
Are Conflict Styles Personality Traits?
Personality dispositions are relevant to conflict preferences, but the connection is more detailed than a rule linking one Big Five trait to one conflict style. A large study by Espinoza, O'Neill, and Donia investigated Big Five factors and 30 personality facets in 11,219 participants. Facet-level models, on average, accounted for approximately one and a half times the variance associated with factor-level models. The findings imply that broad personality categories can conceal more specific dispositions relevant to self-reported conflict approaches (Espinoza, O'Neill, and Donia, 2023).
Such results show statistical associations between measured personality characteristics and measured style tendencies. They do not imply that a person is destined to compete, accommodate, or avoid, and they cannot establish how that person behaves across all contexts. Incentives, relationship histories, norms, perceived threat, and resources can alter which approach is possible or desirable.
A useful distinction is between preference, capacity, and actual behavior. Someone might prefer collaboration, have the capacity to state an independent position, and nevertheless avoid a particular dispute because the circumstances make engagement unsafe. Calling that pattern a contradiction would confuse three separate questions.
Cultural Norms, Identity, and Social Context
Conflict management is embedded in cultural expectations concerning dignity, respect, privacy, directness, hierarchy, obligation, and relational continuity. Some communities place particular value on indirect expressions or on protecting a person's social standing; others expect a direct verbal statement of disagreement. The same pause, concession, or firm refusal may be interpreted differently depending on shared norms and the immediate relationship.
A meta-analysis by Cardon and Okoro examined cultural propositions in face-negotiation theory and found a more differentiated picture than simple stereotypes about individualist and collectivist societies. Their review describes face-related behavior as variable across cultures and relationships, cautioning against reducing people to national categories (Cardon and Okoro, 2010). Other synthesis work has identified patterns by culture, gender, and organizational role but likewise involves heterogeneous instruments and samples, making universal predictions inappropriate (Holt and DeVore, 2005).
Communication disability, language proficiency, sensory environments, neurodevelopmental differences, and discrimination can also shape a participant's accessible options. Direct gaze, rapid conversational response, extensive verbal disclosure, and spontaneous emotional display are culturally and individually variable. These features should never be used as universal criteria for competence or cooperativeness. A person's use of written communication, processing time, an interpreter, or an accommodation may strengthen participation rather than diminish it.
Power matters in every setting, including families, schools, workplaces, and institutional decision-making. A subordinate may accommodate to avoid retaliation, or someone financially dependent may appear to accept a compromise without realistic freedom to reject it. The behavior must be understood within the distribution of risks and available choices.
Power, Coercion, and the Safety Boundary
Ordinary interpersonal conflict involves people whose positions may differ but who have some meaningful capacity to decide whether and how to engage. Coercive control, credible threats, stalking, harassment, economic restriction, and violence change the problem fundamentally. Advice that presumes symmetric responsibility or treats both parties' demands as equally negotiable can expose a targeted person to additional harm.
A request to collaborate is appropriate only when participation can be voluntary, boundaries can be respected, and sharing information does not create a predictable danger. A pressured concession is not evidence that accommodating worked well. A threatening ultimatum is not an ordinary instance of competing. A proposed compromise over physical safety or freedom from abuse lacks the legitimacy of a routine trade-off.
Professional discussion of mediation in cases involving intimate partner violence emphasizes individualized screening, informed participation, safety, and the possibility that a given process is unsuitable. This complex professional literature does not support a universal rule that every dispute should move toward face-to-face dialogue or mediation (American Bar Association, Intimate Partner Violence and Mediation).
For someone facing threats or abuse, appropriate priorities can include physical safety, confidential support, documenting incidents when safe, obtaining legal or specialist advice, and declining joint discussions. The five-style model remains useful as a description of some observable actions, but it is not a substitute for safety assessment, safeguarding practice, or legal protection.
Choosing a Conflict Style According to the Situation
Start with the actual problem and its constraints
A disciplined decision begins by identifying the issue that requires action. Is it a misunderstanding, a disputed allocation, a recurring unmet obligation, a difference in values, or a safety concern? Which interests can change, and which obligations are fixed by law, health, consent, or responsibility? What resources exist? Which participants will carry the costs? These questions determine whether negotiation is appropriate at all.
Next, assess timing and interactional capacity. A heated argument might call for an agreed pause. An urgent threat might require immediate protective action. A minor preference difference might be resolved by voluntary accommodation. A recurring shared responsibility may justify sustained collaborative work. A time-limited decision with reasonably balanced stakes may invite compromise. A clear duty or nonnegotiable entitlement may require a firm stance.
Finally, separate the mode from the quality of its execution. Collaboration can be domineering in practice if only one participant frames the options. Competition can be procedurally respectful if a boundary is stated clearly. Avoidance can be responsible if it includes a realistic return plan. A voluntary accommodation can be an expression of care. The evidence supports selecting responses based on identifiable conditions rather than attempting to maintain a permanently high score on one preferred style.
When a pause is the best first move
Use a temporary pause when arousal, confusion, or exhaustion makes immediate discussion unlikely to be productive and when postponement is safe. Specify how the issue will be revisited, who needs information, and what interim arrangements apply. An agreed pause respects both the need for regulation and the importance of the unresolved issue. Repeated indefinite delay is a different pattern.
When yielding is appropriate
Yield when the other party has a substantially greater legitimate stake, when one's own cost is minor, or when an informed concession advances a valued relationship or reciprocal arrangement. The decisive distinction is whether the person can make the choice freely and knows what they are giving up. A history of coerced yielding signals a need to examine power rather than cultivate more accommodation.
When firmness is required
Compete for a clear outcome when a requirement cannot appropriately be traded away, or when a decision deadline makes prolonged joint exploration impossible. Explain the rationale, the limits of discretion, and the available procedural safeguards. Rejecting an unfair or unsafe demand may be both firm and respectful.
When a limited bargain works
Compromise when a negotiable difference can be allocated fairly, both parties can consent, information is sufficiently clear, and the stakes do not justify extensive exploration. An explicit review point makes practical sense when outcomes are uncertain. Do not confuse equal concessions with equitable costs: losing one hour of sleep may affect two people differently.
When deeper joint analysis is worth the effort
Collaborate when interests are complex, multiple resources or issues can be rearranged, ongoing coordination is important, and parties have enough trust and agency to share accurate information. Define needs, brainstorm options before deciding, test feasibility, record responsibilities, and establish a later assessment. A successful collaborative process should make sense in concrete outcomes rather than merely sound agreeable.
Case Examples: How the Same Label Changes With Context
A disagreement between friends
Two friends want different weekend activities. If the choice matters much more to one, accommodation may be generous. If both have genuine constraints, compromise through alternating plans may work. If one wants connection while the other needs a quiet setting, collaboration can uncover alternatives that satisfy the underlying purposes. This case involves minor stakes and relative freedom of choice, so it cannot serve as an ethical template for high-risk disputes.
An urgent procedural decision
A technician identifies a safety hazard before a public event. Other organizers want to proceed to avoid delays. Firmly insisting on the mandated safety procedure may look competitive within the five-style taxonomy. Evaluating the response requires attention to duty and foreseeable harm. Negotiating a midpoint between safety compliance and noncompliance is not necessarily responsible.
Repeated household responsibility
Two family members disagree about who performs an ongoing task. One always agrees to take on more work, leaving the appearance of harmony while experiencing increasing burden. Naming their needs can change the bargaining process. A provisional compromise may distribute duties; collaboration may reveal differences in physical capacity, work schedules, finances, and support networks. The useful measure is whether the arrangement is equitable and sustained, not whether the participants use the same word for their conflict style.
An employee facing retaliation
An employee believes a manager's demand violates a policy but fears losing work if they object. Accommodating or avoiding may describe the visible response while concealing a constrained choice. The practical issue is the availability of protected reporting, external advice, evidence preservation, and institutional accountability. Applying the same communication exercise used for equals would neglect the central causal factor.
A relationship with coercive control
A person is threatened after questioning a partner's restrictions on communication or money. This situation is not properly framed as two equally valid conflict styles needing adjustment. The relevant concerns are autonomy, danger, and access to confidential support. Pressuring the targeted person into direct negotiation or mutual concession can increase risk.
How to Expand a Conflict-Management Repertoire
The most useful learning goal is flexible and accountable selection among legitimate responses. Begin by describing several recent disagreements without immediately assigning personal labels. Record the issue, the participants' stakes, the available choices, who held decision-making power, the emotional climate, the response taken, and what happened afterward. A repeated pattern matters more than a single score.
Next, identify alternatives that would have been realistically possible. If a person routinely delays low-risk conversations, they might practice scheduling a discussion with a clear topic and start time. If they tend to concede without assessing costs, they might rehearse describing their actual constraint. If they insist on a first proposal too quickly, they might practice asking about the other participant's relevant needs. These are targeted behavioral experiments, not prescriptions to become uniformly agreeable.
Learning is stronger when rehearsal includes feedback from actual behavior rather than a declaration of preferred identity. Role-play can help a person test wording, but everyday transfer requires follow-up: did the conversation occur, did the other participant understand the proposal, was the outcome feasible, and were commitments honored? The communication evidence base varies by setting, and training success measured immediately after an exercise should not automatically be equated with durable everyday improvement.
A deliberate repertoire also includes protective withdrawal and appropriate escalation to authorities or independent professionals. The model should never train a person to remain in unsafe conversations to demonstrate emotional maturity. The right skill may be recognizing that negotiation is not the relevant task.
Limits of the Scientific Evidence
The five-style framework has substantial historical influence and a coherent descriptive structure. Those strengths do not make every assumption equally well established. The existence of five teachable labels is well documented; the proposition that precisely five stable psychological categories exhaust conflict behavior is contested. Experimental and observational studies often employ different conceptualizations, measurement instruments, and outcomes, limiting direct synthesis (Guerrero, 2020; Thomas and Kilmann, 1978).
Self-report instruments can be affected by social desirability, forced-choice interdependence, recall limitations, contextual framing, and the role of the other participant. Sample-specific norms and associations may not transfer across cultures or institutions. A large survey can estimate links among questionnaire scores without establishing that deliberately increasing a score will improve a relationship.
Outcome evidence is often indirect. Meta-analyses of negotiation and relationship interaction show that social motivation, hostility, problem-solving, and context matter, but these studies do not constitute a randomized comparative trial in which the five styles are assigned as complete interventions across ordinary lives. Strong claims that collaborating is universally optimal, avoiding is universally harmful, or compromising always divides benefits equally exceed the literature.
The most scientifically defensible application is therefore conditional: describe the interaction accurately, identify constraints and plausible mechanisms, investigate outcomes, and revise behavior when evidence warrants. The framework is valuable because it makes options explicit, while empirical evaluation requires more than a memorable typology.
Evidence Strength: Established, Preliminary, Limited, and Contested
Established: the five modes are an influential descriptive classification; the two-dimension structure is an explicit theoretical foundation; major questionnaire instruments exist; conflict responses vary with context; and social-desirability and measurement concerns have been empirically investigated (Kilmann and Thomas, 1977; Rahim, 1983).
Supported but context-dependent: constructive problem solving, motivation, and relational behavior are associated with relevant outcomes in studies of negotiation and couples. The size and meaning of these associations depend on design, population, comparison, and setting (De Dreu, Weingart, and Kwon, 2000; Woodin, 2011).
Limited: evidence that changing one person's self-reported five-style profile, by itself, produces lasting improvement across unrelated types of interpersonal conflict. Direct causal and generalization evidence is thinner than educational accounts sometimes suggest.
Contested: the completeness of a five-mode taxonomy, the degree to which scores represent stable traits, and any universal hierarchy declaring a single best style. Work identifying a sixth interpersonal pattern and finer-grained trait associations provides reasons to test these questions rather than assume their answers (Guerrero, 2020; Espinoza, O'Neill, and Donia, 2023).
Frequently Asked Questions
What are the five conflict management styles?
The five commonly taught conflict management styles are avoiding, accommodating, competing, compromising, and collaborating. They differ in how strongly a person pursues their own concerns and the concerns of others during a disagreement. The five styles are descriptive strategies or tendencies, not diagnoses or permanent types.
Which conflict management style is best?
The appropriate response depends on stakes, rights, safety, timing, relational goals, and available options. Collaboration may be valuable when participants can jointly explore interests and construct workable solutions. A firm refusal can be necessary when a basic obligation or boundary is at issue. Temporary avoidance can protect safety or allow better preparation. There is no universally superior style for every conflict.
Is avoiding conflict always unhealthy?
No. Avoidance can be a responsible temporary pause, a decision that a minor issue does not warrant engagement, or protective non-engagement in a dangerous setting. Its consequences depend on the issue, the duration of withdrawal, and whether necessary responsibilities are addressed.
What is the difference between accommodating and compromising?
Accommodating primarily yields to the other party's preference. Compromising involves an exchange in which parties typically adjust or relinquish some demands to reach an acceptable settlement. Either may be voluntary and reasonable or may disguise unequal bargaining power.
How are compromising and collaborating different?
Compromising works with concessions among established demands. Collaborating seeks to clarify underlying interests and develop alternatives that may address more of those interests. Collaboration often requires additional time, information, and reciprocal engagement; it does not guarantee agreement.
Can a person have more than one conflict style?
Yes. People can prefer several styles and use different responses with different people or at different moments in the same dispute. Questionnaires describe reported tendencies, while real behavior depends on circumstances and constraints.
Is the Thomas–Kilmann test a personality or mental-health diagnosis?
No. The TKI is a conflict-response assessment. Its scores can support reflection and training, subject to the instrument's limitations. They do not diagnose a psychological condition, independently determine whether someone is abusive, or predict future conduct with certainty.
Does a collaborative approach work when the other person is threatening or controlling?
Meaningful collaboration requires voluntary participation and safety. Threats, coercion, and violence undermine those requirements. Protective action, independent professional assistance, and safeguarding considerations take precedence over attempts to negotiate through a five-style exercise.
Related Articles
Cluster Communication, Conflict and Social Skills — the research foundation for interpersonal meaning, social competence, and conflict processes.
Assertive Communication: Skills, Examples, and How to Practice — expressing needs and boundaries clearly without equating assertiveness with competition.
Power Dynamics in Relationships: Influence, Dependence, Control, and Conflict — how unequal dependence and control alter the freedom to negotiate.
Silent Treatment: When Withdrawal Becomes Punishment or Control — distinguishing protective pauses from punitive withdrawal.
Relationship Repair After Conflict: How Couples Recover Emotional Closeness — the specific processes of post-conflict repair in intimate relationships.
Interpersonal Conflict in Psychology: Causes, Dynamics and Outcomes — Define interpersonal conflict and distinguish needs, roles, values and power.
Types of Interpersonal Conflict: Interests, Values, Roles and Relationships — Classify types of disagreements with nonclinical contextual examples.
Compromise vs Collaboration: What Works in Different Conflicts — Compares partial concessions, integrative outcomes and constraints.
References
American Bar Association. (n.d.). Intimate partner violence and mediation. American Bar Association.
Cardon, P., and Okoro, E. (2010). A meta-analysis of the cultural propositions about conflict management styles in face-negotiation theory. Journal of Rhetoric, Professional Communication, and Globalization, 1(1), Article 2. Purdue University repository.
De Dreu, C. K. W., and Weingart, L. R. (2003). Task versus relationship conflict, team performance, and team member satisfaction: A meta-analysis. Journal of Applied Psychology, 88(4), 741–749. https://doi.org/10.1037/0021-9010.88.4.741.
De Dreu, C. K. W., Weingart, L. R., and Kwon, S. (2000). Influence of social motives on integrative negotiation: A meta-analytic review and test of two theories. Journal of Personality and Social Psychology, 78(5), 889–905. PubMed.
Espinoza, J. A., O'Neill, T. A., and Donia, M. B. L. (2023). Big Five factor and facet personality determinants of conflict management styles. Personality and Individual Differences, 203, 112029. https://doi.org/10.1016/j.paid.2022.112029.
Guerrero, L. K. (2020). Conflict style associations with cooperativeness, directness, and relational satisfaction: A case for a six-style typology. Negotiation and Conflict Management Research, 13(1), 24–43. https://doi.org/10.1111/ncmr.12156.
Holt, J. L., and DeVore, C. J. (2005). Culture, gender, organizational role, and styles of conflict resolution: A meta-analysis. International Journal of Intercultural Relations, 29(2), 165–196. https://doi.org/10.1016/j.ijintrel.2005.06.002.
Kilmann, R. H., and Thomas, K. W. (1977). Developing a forced-choice measure of conflict-handling behavior: The “MODE” instrument. Educational and Psychological Measurement, 37(2), 309–325. https://doi.org/10.1177/001316447703700204.
Rahim, M. A. (1983). A measure of styles of handling interpersonal conflict. Academy of Management Journal, 26(2), 368–376. https://doi.org/10.5465/255985.
Rahim, M. A. (1986). Referent role and styles of handling interpersonal conflict. The Journal of Social Psychology, 126(1), 79–86. https://doi.org/10.1080/00224545.1986.9713573.
Thomas, K. W., and Kilmann, R. H. (1978). Comparison of four instruments measuring conflict behavior. Psychological Reports, 42(3, Suppl.), 1139–1145. https://doi.org/10.2466/pr0.1978.42.3c.1139.
van de Vliert, E., and Euwema, M. C. (1994). Agreeableness and activeness as components of conflict behaviors. Journal of Personality and Social Psychology, 66(4), 674–687. https://doi.org/10.1037/0022-3514.66.4.674.
Woodin, E. M. (2011). A two-dimensional approach to relationship conflict: Meta-analytic findings. Journal of Family Psychology, 25(3), 325–335. https://doi.org/10.1037/a0023791.